There are two key problems for tuberculosis diagnosis worldwide: inadequate tests and inadequate access to testing. Billions of dollars have been invested in developing novel platforms that are more accurate than sputum smear microscopy. However, tests in the diagnostic pipeline are currently too costly, or require too much infrastructure, to be used in the peripheral health centers that serve 85% of the population in low-income countries.
Project Vision and Strategy
The CellScopeTB is an iPad-based diagnostic system that serves as a fully-automated digital microscopy platform. It is designed to improve the quality of existing microscopy services and expand access to quality-assured TB microscopy for patients in low-resource settings. With DIL support, the research team is developing the software and IT infrastructure for a second-generation diagnostic platform, including: 1) an App that includes a self-calibration routine, auto-focusing and image quality assessment capability, and optimized image analysis algorithm; and 2) a cloud-based infrastructure for storing patient and image data, and communicating results to patients via SMS. The team plans to evaluate the reproducibility, accuracy, impact and cost-effectiveness of the CellScopeTB’s automated digital microscopy in Hanoi, Vietnam.
The research team continues to refine their design-for-manufacturing work and they have developed software solutions for self-calibration and auto-focusing of the device. The goal is to advance CellScope TB to be a fully automated digital fluorescence microscopy platform. They are also refining the image analysis algorithm and have made plans to train and validate the algorithm as well as assess reproducibility of the manufactured device using banked sputum smears and associated clinical data from completed studies in Nigeria and Uganda.
The team recently collected baseline (pre-intervention) data from four participating district health centers in Ha Noi. In launching this study in Vietnam, the team will be working with scale-up partners in the US for manufacturing. They’ve also submitted an abstract describing the pre-intervention data to the Union World Conference on Lung Health.
- Dr. Adithya Cattamanchi, Medicine, UCSF
- Professor Dan Fletcher, Bioengineering, UC Berkeley
- Dr. Payam Nahid, Medicine, UCSF
Relevant Publications and Press
- “Quantitative Imaging with a Mobile Phone Microscope,” by Arunan Skandarajah, Clay D. Reber, Neil A. Switz, and Daniel A. Fletcher, PLOS ONE, May 2014.
- “Mobile Digital Fluorescence Microscopy for Diagnosis of Tuberculosis,” by Asa Tapley, Neil Switz, Clay Reber, J. Lucian Davis, Cecily Miller, John Baptist Matovu, William Worodria, Laurence Huang, Daniel A. Fletcher, and Adithya Cattamanchi,Journal of Clinical Microbiology, April 2014.
- “Automated Tuberculosis Diagnosis Using Fluorescence Images from a Mobile Microscope,” by Jeannette Chang, Pablo Arbeláez, Neil Switz, Clay Reber, Asa Tapley, J. Lucian Davis, Adithya Cattamanchi, Daniel Fletcher, and Jitendra Malik,Medical Image Computing and Computer-Assisted Intervention, May 2012.